vBond Prod Interface speed

[cloudlogics]

I see high packet drops on vBond prod interface (on-prem deployment). Using ethtool I can see that interface is negotiated at 10/full.

VBONDXX:~$ ethtool ge0_0
Settings for ge0_0:
Supported ports: [ ]
Supported link modes: Not reported
Supported pause frame use: No
Supports auto-negotiation: No
Advertised link modes: Not reported
Advertised pause frame use: No
Advertised auto-negotiation: No
Speed: 10Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 0
Transceiver: internal
Auto-negotiation: off
MDI-X: Unknown

Is there a way to change speed to 1000/Full? I have multiple vBonds across multiple DCs and they all show the 10/full. Other controllers (vSmart/vManage) show 1000/full. There is no speed command available under the interface itself neither does it allow me to change it via the shell using ethtool getting "Operation not permitted" message.

Underlay VM infra is configured fine. Some vBonds and vSmarts share the same host.

[Kanan Huseynli]

Hi,

what does "show interface" show? Also you can use iperf in vbond (vedge-cloud) :

tools iperf vpn 0 options "-c [server_ip] [additional parameters]"

In my vbond in vshell linux command shows 10Mbps, however show interface shows 1000/Full:

vedge# sh interface

IF IF IF TCP
AF ADMIN OPER TRACKER ENCAP SPEED MSS RX TX
VPN INTERFACE TYPE IP ADDRESS STATUS STATUS STATUS TYPE PORT TYPE MTU HWADDR MBPS DUPLEX ADJUST UPTIME PACKETS PACKETS
------------------------------------------------------------------------------------------------------------------------------------------------------------------------
0 ge0/0 ipv4 172.20.1.1/24 Up Up NA null transport 1500 00:0c:29:3f:1d:c5 1000 full 1416 8:20:47:54 3157318 4053359

And, finally, how did you check drops on device?

[cloudlogics]

Thanks
Show interface does show 1000/full. I used "show int detail ge0/0" to see packet drops. I am not sure if this is what is causing the packet drops but I have multiple vbonds and they all show same behaviour. May have to go TAC route. 

[Kanan Huseynli]

I think packet drops can also be implicit access-list. Check for utilization and also do you have - feel actual issue? Try to do contiguous ping for vpn0 tunnel interface, also with higher packet size to see packet loss %. 

By the way, sometimes outputs shows cosmetic errors or statistical information (which is incorrect in reality).

And does vbond behind firewall with 1:1 nat or just public IP on interface?

If all are affected and all have the same release, it can also be bug.